ACI 355.

2 – SEISMIC TESTING OF POST-INSTALLED CONCRETE AND

MASONRY ANCHORS IN CRACKED CONCRETE

GILBERT BALBUENA

Simpson Strong-Tie Australia

In January 2002, the American Concrete Institute (ACI) issued a new standard, ACI355.2-
01, which prescribed a comprehensive testing program to ascertain design parameters for
post-installed adhesive anchors and/or mechanical anchors used in cracked or uncracked
concrete. Clause 17.5.6 NZS 3101: Part 1:2006 or ACI 318 Appendix D may also be used
for post-installed mechanical anchors that have passed the qualification test stipulated in
ACI 355.2. Post-installed mechanical anchors intended to resist seismic actions shall have
passed the seismic test of ACI 355.2”

ACI 355.2 applies to post-installed mechanical anchors intended for use in structural
applications addressed by ACI 318 and subjected to static or seismic loads in tension, shear,
or combined tension and shear. It does not apply to anchors loaded in compression, or to
anchors subjected to long-term fatigue loading. Anchors meeting the requirements of ACI
355.2 are expected to sustain their design loads (in tension, shear, and combined tension
and shear) while providing adequate stiffness.

ACI355.2 prescribes testing and evaluation requirements for post-installed mechanical

anchors (torque-controlled expansion anchors, displacement-controlled expansion anchors
and undercut anchors) with nominal diameter of 6mm or larger intended for use in concrete
designed under the provisions of ACI 318. Criteria are prescribed to determine whether
anchors are acceptable for use in uncracked concrete only or in cracked as well as
uncracked concrete. Criteria are prescribed to determine the performance category for each
anchor. The anchor performance categories are used by ACI 318 to assign reduction factors
and other design parameters.

CRACKED CONCRETE ANCHOR

“CRACKED CONCRETE” refers to a situation where a crack passes directly through the
hole in which the anchor is installed. A “cracked concrete anchor,” refers to an anchor that
has been tested and qualified for applications where cracking could be expected.

Many types of anchors will not work well, or at all, when tested in cracked concrete, but
some anchors are specifically designed and tested to perform in cracks. Since there have
not been any documented real-life anchor failures resulting from cracked concrete in the
past, there has been debate as to whether or not it is necessary to consider ‘cracked
concrete’ in anchor testing and design. Regardless, anchor testing in cracked concrete has
officially been adopted by the International Building Code and the code requires these test
standards be met, in order to be used on jobsites.

However, since post-installed anchor shapes and designs vary greatly based on product
type and manufacturer, post-installed anchors MUST be tested to the new standard in order
to determine their behavior in a crack.

An anchor should be tested and qualified in cracked concrete anytime it is intended to be

used in an application where the concrete is expected to crack at some point throughout its
use.
For instance, the Clause 17.5.6 NZS 3101: Part 1:2006 states that cracked concrete anchors
must be used anytime the concrete member will be used to resist seismic forces.

In general anchor testing is performed in cracks to try and simulate the conditions that could
happen at some point throughout its use. ACI355.2 allows manufacturers to test for
recognition into concrete under the following conditions: uncracked concrete, cracked and
uncracked concrete, or cracked/uncracked concrete and seismic. The codes do not permit
an anchor be used in most seismic regions if it has not been tested to the last condition.

The tests required to earn recognition in the latter condition requires the anchor to pass tests
that are performed in cracked concrete with a moving load and/or moving crack width. Not
only must the anchor stay in the hole and resist a minimum load, it must also be within
acceptable displacement limits. This is a very difficult test to pass and often only anchors
that have been intentionally designed to perform in this test will pass it.

WHAT CAUSES CONCRETE TO CRACK?

Common conditions that cause cracking:

• Concrete in tension – such as in the underside of a slab
• Seismic Areas
• Other factors that contribute to cracking include:
– External short term loads (such as high winds)
– Temperature variations
– Shrinkage during curing
– Improper design

Concrete is prone to cracking in areas called the ‘tension zone’. Any area of a member that
has tension is called a tension zone, however, that does not necessarily mean that cracking
will occur in every tension zone.

Concrete is strong in compression but relatively weak in tension, and even though it is weak
it still has some ability to carry tension. The tension strength is approximately 10% of the
compression strength.

The material property to measure concrete’s tension capacity is called the Modulus of
Rupture (fr). The Code gives an equation for calculating this Modulus as f r = 7.5 times the
square root of (f’c), where f’c is the compressive strength of the concrete.

In order for cracking to occur, the tension stress in the tension zone must exceed the
Modulus of Rupture.

Let’s look at an example to help show how to locate tension zones and where to expect
cracking.
This concrete beam is being supported by the two yellow shapes. As this beam is loaded by
any number of sources, it begins to deflect as shown by the blue dashed line. The area at
the top of the concrete beam is in compression, as it is being squeezed together. The area
at the bottom of the beam is being stretched apart and reflects a “tension zone”. It is in the
tension zone that cracking could be expected, but only if the tension force exceeds the
concrete’s tension capacity.

As you can imagine, when the unreinforced concrete does crack, it no longer has any
tension carrying capacity. Engineers know that concrete will crack if enough tension force is
developed and as a result they specify that the steel rebar be installed in tension zones of
concrete members. Steel works well in tension and will carry the tension force in areas
where the concrete alone would most likely fail. If we were to take a slice through the mid-
span of the previous beam that was used as an example, you would see a lot more rebar in
the bottom of the beam than in the top. Again, the top of the beam is in compression and
concrete is strong in compression so it doesn’t need as much reinforcement from rebar. But
the bottom of the beam is in tension and since concrete is weak in tension, the rebar will be
used to carry all the tension once the concrete cracks.

You may hear that rebar prevents concrete from cracking so there’s no need to use cracked
concrete anchors if rebar is in the concrete. This is not necessarily true.

Rebar is used in concrete for both:

1.) tension strength, and
2.) to minimize cracks and deflections,

Unless rebar is intentionally over-designed to do so, a normal concrete member reinforced

with rebar will still have some cracking in it.

The new anchor test standard ACI 355.2, DOES assume there will be some rebar in the
cracked concrete zone to control the crack widths to a Code limit. It is this same Code crack
width limit that is used to establish the crack widths that are tested to in the new ‘cracked
concrete’ test criteria.

EXAMPLES OF TENSION ZONES

Tension Zones and cracks are not always located on the “underneath side” of a horizontal
concrete member.

In addition, cracks do not only occur in beams. They can occur in footings, columns, and any
other structural member where the tension load exceeds the concrete’s ability to carry that
tension load.

When designing concrete anchorages, designers are now being required to consider, among
other things, whether conditions exist that may cause the concrete to crack. If it’s determined
such conditions do exist, anchors designed and tested for use in cracked concrete must be
specified. If it’s determined there is no risk of concrete cracking, the designer may choose to
specify anchors that meet the 2006 IBC requirements for use in un-cracked concrete.
The Cast in place anchors tested, all conformed to a specific size and shape, as defined by
an industry standard known as ANSI 18.2.1 and 18.2.6. As a result, if engineers design
using a cast-in-place anchor that meets this size and shape standard, it does not need to be
tested and qualified to the new anchor standard, it is basically “grandfathered” into the
existing test database.

However, since post-installed anchor shapes and designs vary greatly based on product
type and manufacturer, post-installed anchors MUST be tested to the new standard in order
to determine their behavior in a crack.

Whether or not this cracking is expected can be calculated and is indicated by the Code. For
instance, the Code states that cracked concrete anchors must be used anytime the concrete
member will be used to resist seismic forces. Unlike gravity load which always acts
downward, a seismic load can come from any direction –from either side or even up and
down. Accordingly, the Code requires that a cracked concrete anchor be used for regions of
medium to high seismicity. Even if the anchor is located in a region where cracking is not
expected under gravity loads the Code still requires it be tested in a crack in these regions.

Cracked concrete anchors are also required in areas called tension zones. Although
cracking will occur from applied structural loads and thermal effects (such as temperature
and shrinkage cracks), the cracks covered by the new Code and test Standard are intended
to address those resulting only from applied loads. In general anchor testing is performed in
cracks to try and simulate the conditions that could happen at some point throughout its use.

Much research has gone into anchor performance in cracks. Cast-in-place anchors were
tested in cracks and consistently showed about a 25% loss from testing done in uncracked
concrete. Many types of post-installed anchors were tested, wedge anchors, drop-in
anchors, and adhesive anchors. And they all showed a loss in performance when tested in
a crack. The expansion anchors tested, showed a loss of about 40% in a crack, drop-in
anchors showed about an 80% loss and adhesive anchors showed a loss of 60%.

ACI 355.2 – QUALIFICATION OF POST-INSTALLED MECHANICAL ANCHORS

Testing and evaluation of anchors are performed or witnessed by an independent testing

and evaluation agency accredited under ISO/IEC 17025 by a recognized accreditation body
conforming to the requirements of ISO/IEC 17011. Also the testing shall be certified by a
licensed engineer employed or retained by the independent testing and evaluation agency.
Reference and reliability tests shall all be performed by an independent testing and
evaluation agency. Service-condition tests performed by the manufacturer shall not be more
than 50% of the total number of tests. Tests done by the manufacturer shall be witnessed by
an independent testing laboratory or engineer meeting the requirement of ACI355.2 Chapter
12. Tests done by the manufacturer shall only be considered in the evaluation if the results
are statistically equivalent to those of the independent testing and evaluation agency.
Testing Sequence (refer to flowchart)
1. Identification test – to evaluate the anchors compliance with the critical
characteristics.
2. Reference tests – to establish baseline performance against which subsequent tests
are to be compared.
3. Reliability tests – to confirm the reliability of the anchor under adverse installation
procedures and long-term use.
4. Service-condition tests – to evaluate the performance of the anchor under expected
service conditions.

Test program for evaluating anchor systems for use in uncracked concrete
Concrete Member Drill bit Minimum
Test numberSection Purpose Description 1
Strength Thicknessdiameters ample size n
Reference tests
1 7.2 Reference test in uncracked low-strength concrete Tension - single anchor with no edge influence Low ≥ hmin dm 5
2 7.2 Referecne test in uncracked high-strength concrete Tension - single anchor with no edge influence High ≥ hmin dm 5
Reliability tests
2
3 8.2 Sensitivity to reduced installation effort Tension - single anchor with no edge influence Varies with anchor type ≥ hmin dm 5
4 8.3 Sensitivity to large hole diameter Tension - single anchor with no edge influence Low ≥ hmin dm 5
5 8.4 Sensitivity to small hole diameter Tension - single anchor with no edge influence High ≥ hmin dm 5
6 8.5 Reliability under repeated load Repeated tension - single anchor with no edge influence, residual capacity Low ≥ hmin dm 53
Service-condition tests
7 9.2 Verification of full concrete capacity in corner with two edges located at 1.5h ef Tension - single anchor in corner with two edges located at 1.5h ef Low hmin dm 4
8 9.3 Minimum spacing and edge distance to preclude splitting on installation High installation tension (torque or direct) - two anchors near edge Low hmin dm 5
4
9 9.4 Shear capacity of anchor steel Shear - single anchor with no edge influnce Low ≥ hmin dm 5

1
Minimum sample size for each diameter, unless otherw ise noted.
2
Drill bit diameter for undercuts are specified in Table 5.6
3
Tests are not required for each diameter. Test smallest, middle and the largest anchor diameter.
4
Required only for anchors w hose cross-sectional area, w ithin five anchor diamters of the shear failure plane, is less than that of a threaded bolt of the same nominal diameter as the anchor, or for sleeved anchors w hen shear capacity of the sleeve w ill be considered.

Test program for evaluating anchor systems for use in cracked and uncracked concrete
Crack opening Concrete Member Drill bit Minimum
Test numberSection Purpose Description 1
width (mm) Strength Thicknessdiameters ample size n
Reference tests
1 7.2 Reference test in uncracked low-strength concrete Tension - single anchor with no edge influence - Low ≥ hmin dm 5
2 7.2 Referecne test in uncracked high-strength concrete Tension - single anchor with no edge influence - High ≥ hmin dm 5
3 7.2 Reference test in low-strength, cracked concrete Tension - single anchor with no edge influence 0.3 Low ≥ hmin dm 5
4 7.2 Referecne test in high-strength, cracked concrete Tension - single anchor with no edge influence 0.3 High ≥ hmin dm 5
Reliability tests
2
5 8.2 Sensitivity to reduced installation effort Tension - single anchor with no edge influence 0.3 Varies with anchor type ≥ hmin dm 5
6 8.3 Sensitivity to crack width and large hole diameter Tension - single anchor with no edge influence 0.5 Low ≥ hmin dm 5
7 8.4 Sensitivity to crack width small hole diameter Tension - single anchor with no edge influence 0.5 High ≥ hmin dm 5
8 8.6 Test in cracks whose opening width is cycled Repeated tension - single anchor with no edge influence, residual capacity0.1 to 0.3 Low ≥ hmin dm 53
Service-condition tests
9 9.2 Verification of full concrete capacity in corner with two edges located at 1.5h ef Tension - single anchor in corner with two edges located at 1.5h ef - Low hmin dm 4
10 9.3 Minimum spacing and edge distance to preclude splitting on installation High installation tension (torque or direct) - two anchors near edge - Low hmin dm 5
4
11 9.4 Shear capacity of anchor steel Shear - single anchor with no edge influnce - Low ≥ hmin dm 5
12 5 9.5 Seismic tension Pulsating tension, single anchor, with no edge influence 0.5 Low ≥ hmin dm 5
5
13 9.6 Seismic shear Alternating shear, single anchor, with no edge influene 0.5 Low ≥ hmin dm 5

1
Minimum sample size for each diameter, unless otherw ise noted.
2
Drill bit diameter for undercuts are specified in Table 5.6
3
Tests are not required for each diameter. Test smallest, middle and the largest anchor diameter.
4
Required only for anchors w hose cross-sectional area, w ithin five anchor diamters of the shear failure plane, is less than that of a threaded bolt of the same nominal diameter as the anchor, or for sleeved anchors w hen shear capacity of the sleeve w ill be considered.
5
These tests are optional
Concrete strength for testing shall be within these ranges:
Low-strength concrete: 17 to 28 MPa
High-strength concrete: 46 to 60 MPa

Anchors installation shall be as per manufacturer installation instructions. Drill holes shall
perpendicular (within a tolerance of +- 6 degrees) to the surface of the concrete member.
Required diameter of carbide hammer-drill bits shall be as per the table below.

Test members shall be permitted to contain reinforcement to allow handling, the distribution
of loads transmitted by the test equipment, or both. Placements of the reinforcement so that
the capacity of the tested anchors are not affected.

The crack width shall be approximately uniform throughout the member thickness. The
thickness of the test member shall not be less than 1.5hef but at least 100mm. To control the
location of the cracks and to ensure that the anchor are installed to the full depth in the
cracks, cracks inducers shall be permitted to be installed in the member provided that they
are not situated so as to influence the test results.

DIFFERENT TYPES OF CRACKED CONCRETE TESTING

There are essentially two types of cracked concrete tests: static-crack and moving-crack. In
static-crack testing, the anchor is tested in the crack while the crack is held to a specified
width. In moving-crack testing, the anchor is tested in a crack which is cycled between two
widths.

Engineers first create a hairline cracked in the concrete member using an established
method. Then, they drill a hole perpendicular to the surface of the concrete member,
effectively bisecting the crack, and thereby positioning the axis of the anchor in the plane of
the crack. An anchor is installed and set pursuant to the manufacturer's instructions (as
modified by the criteria). The subsequent testing is conducted as follows:

For static crack tests: The crack is opened to a specified width (either 0.012 or 0.020 inch)
and held. Then the anchors are tested in tension or shear, while the crack width is
monitored.

For moving crack tests: The crack width is cycled between two widths, starting with 0.004
and 0.012 inch, while a static load is applied to the anchor. The crack is cycled one thousand
times while any resulting displacement of the loaded anchor is recorded. After the cycling is
completed, the crack is held static, and the anchor is pulled to failure.

"ADVERSE" CONDITIONS FOR DIFFERENT TYPES OF ANCHORS

The adverse conditions will vary depending on the type of the anchor, and in some cases,
the preferences of the manufacturer. For mechanical anchors, the adverse conditions
include drill holes at maximum tolerance or minimum tolerance and at lower installation
torque (where applicable). For adhesive anchors, the adverse conditions include installation
in saturated concrete, submerged in water, etc. In addition, testing for effects of freeze-thaw,
sustained load and installation direction are commonly required.
 Summary of test requirements

Test number Purpose Description Requirments

Reference tests
1 Reference test in uncracked low-strength concrete Tension - single anchor with no edge influence
Coefficient of variance of the
2 Referecne test in uncracked high-strength concrete Tension - single anchor with no edge influence
ultimate tension load in any test
3 Reference test in low-strength, cracked concrete Tension - single anchor with no edge influence
series < 15%
4 Referecne test in high-strength, cracked concrete Tension - single anchor with no edge influence
Reliability tests
5 Sensitivity to reduced installation effort Tension - single anchor with no edge influence
Coefficient of variance of the
6 Sensitivity to crack width and large hole diameter Tension - single anchor with no edge influence
ultimate tension load in any test
7 Sensitivity to crack width small hole diameter Tension - single anchor with no edge influence
series < 20%
8 Test in cracks whose opening width is cycled Repeated tension - single anchor with no edge influence, residual capacity
Service-condition tests
Capacity of the anchor with two
edge distance of 1.5hef shall be
statistically equivalent to the
9 Verification of full concrete capacity in corner with two edges located at 1.5h ef Tension - single anchor in corner with two edges located at 1.5h ef
capcity of the reference tests
performed with no edge
influence.
5% fractile of the maximum
recorded torque calculated > the
10 Minimum spacing and edge distance to preclude splitting on installation High installation tension (torque or direct) - two anchors near edge
smaller of 1.7Tinst or 1.0Tinst +
135Nm
4
11 Shear capacity of anchor steel Shear - single anchor with no edge influnce
All anchors in the test series
shall complete the simulated
12 Seismic tension Pulsating tension, single anchor, with no edge influence
seismic-tension load history
specified
All anchors in the test series
shall complete the simulated
13 Seismic shear Alternating shear, single anchor, with no edge influene
seismic-tension load history
specified

ESTABLISHING ANCHOR CATEGORIES

In order to established the anchor category, for each combination of anchor diameter and
embedment depth, the ratio of the characteristic capacity Nb,r in each reliability test to the
characteristic tension capacity Nb,o in the corresponding reference test. Using the smallest
ratio Nb,r/Nb,o from the reliability tests, establish the anchor category as per the table below.

Establishm ent of anchor categories

Smallest ratio of characteristic capacities Anchor category
0.80 ≤ Nb,r / Nb,o 1
0.70 ≤ Nb,r / Nb,o < 0.80 2
0.60 ≤ Nb,r / Nb,o < 0.70 3
If ≤ Nb,r / Nb,o < 0.60 Anchor is unqualified

CONCLUSION

So now that we know testing shows anchor performance is in fact affected by cracks in the
concrete. Anchor should be tested & qualified for cracked concrete anytime it is intended to
be used in an application where the concrete is expected to crack at some point throughout
its use.

For instance, the NZS3101 says that Cracked concrete anchors must be used anytime it will
be used to resist seismic forces. Cracked concrete anchors can also be required in areas
called tension zones.

REFERENCES

ACI 355.2-07 Qualification of Post-Installed Mechanical Anchors in Concrete and

Commentary (Reported by ACI Committee 355), American Concrete Institute

ACI 355.4-11 Qualification of Post-Installed Adhesive in Concrete and Commentary

(Reported by ACI Committee 355), American Concrete Institute

Anchorage in Concrete Construction, R.Eligehausen, R. Mallee, J. Silva (2006)